Merging sequential e-values via martingales

DOI10.1214/24-EJS2228MaRDI QIDQ6200906zbMATH OpenOpenAlexFDO

Authors Vladimir Vovk, Ruodu Wang

Publication date 25 March 2024

Published in Electronic Journal of Statistics (Search for Journal in Brave)

Copyright license Creative Commons Attribution 4.0 International

Full work available at URL https://arxiv.org/abs/2007.06382, https://projecteuclid.org/journals/electronic-journal-of-statistics/volume-18/issue-1/Merging-sequential-e-values-via-martingales/10.1214/24-EJS2228.full

zbMATH Keywords

admissibilityanytime validitybetting scorese-processesmerging functions

Mathematics Subject Classification ID

Martingales with discrete parameter (60G42) Admissibility in statistical decision theory (62C15) Nonparametric hypothesis testing (62G10)

Abstract: We study the problem of merging sequential or independent e-values into one e-value for statistical decision making. We describe a class of e-value merging functions via martingales, and show that all merging methods for sequential e-values are dominated by such a class. In case of merging independent e-values, the situation becomes much more sophisticated, and we provide a general class of such merging functions based on reordered test martingales.

Cites work

• A New Interpretation of the von Mises' Concept of Random Sequence
• Admissible ways of merging \(p\)-values under arbitrary dependence
• E-values: calibration, combination and applications
• Game-theoretic foundations for probability and finance
• scientific article; zbMATH DE number 1478492 (Why is no real title available?)
• scientific article; zbMATH DE number 3205709 (Why is no real title available?)
• scientific article; zbMATH DE number 3034029 (Why is no real title available?)
• Sum of arbitrarily dependent random variables
• Test martingales, Bayes factors and \(p\)-values
• The Kleene Hierarchy Classification of Recursively Random Sequences
• Time-uniform, nonparametric, nonasymptotic confidence sequences
• Universal inference
• Valid sequential inference on probability forecast performance

Cited in

(4)

• True and false discoveries with independent and sequential e-values
• On the optimality of coin-betting for mean estimation
• The only admissible way of merging arbitrary e-values
• Improved thresholds for e-values

This page was built for publication: Merging sequential e-values via martingales

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6200906)